OECD DEVELOPMENT CENTRE · 2016. 3. 29. · OECD DEVELOPMENT CENTRE 94 rue Chardon-Lagache 75016 Paris, France Tel.: (33-01) 45.24.82.00 Fax: (33-01) 45.24.79.43 DEVELOPMENT CENTRE

POLICY BRIEF No. 16

OECD DEVELOPMENT CENTRE

Helmut Reisen

by

AFTER THE GREAT ASIAN SLUMP:TOWARDS A COHERENT APPROACH

TO GLOBAL CAPITAL FLOWS

• The unprecedented withdrawal of foreign private capital from Asia, morethan 10 per cent of GDP in the crisis countries, confronts them with atransfer problem. Creditor governments should induce their home banksinto financial rescue operations to reduce moral hazard in private-sectorlending, and to encourage Asia’s recovery.

• The resolution of Asia’s domestic debt overhang must be the overridingpolicy concern for Asia’s governments; paying the inevitable fiscal costin Asia’s restructuring process requires tax-base broadening, supportedby easy monetary policy.

• Progress towards a less crisis-prone international financial system willhinge on how to correct the excessive risk taking by banks. Regulatorydistortions through the Basle Accord which bias bank lending towardsthe short term should be corrected.

• Developing countries should strengthen bank and non-bank balancesheets and raise the quality of inflows; Chile-type regulatory measures,however, will only be effective in an appropriate policy context.

POLICY BRIEF No. 16

AFTER THE GREAT ASIAN SLUMP:TOWARDS A COHERENT APPROACH

TO GLOBAL CAPITAL FLOWS

by

Helmut Reisen

© OECD 1999

THE OPINIONS EXPRESSED AND ARGUMENTS EMPLOYED IN THIS PUBLICATION ARE

THE SOLE RESPONSIBILITY OF THE AUTHOR AND DO NOT NECESSARILY REFLECT

THOSE OF THE OECD OR OF THE GOVERNMENTS OF ITS MEMBER COUNTRIES

✒

OECDDEVELOPMENT CENTRE94 rue Chardon-Lagache

75016 Paris, FranceTel.: (33-01) 45.24.82.00 Fax: (33-01) 45.24.79.43

DEVELOPMENT CENTRE POLICY BRIEFS

In its research activities, the Development Centre aims to identify andanalyse problems whose implications will be of concern in the nearfuture to both Member and non-Member countries of the OECD. Theconclusions represent a contribution to the search for policies to dealwith the issues involved.

The Policy Briefs deliver the research findings in a concise and accessibleway. This series, with its wide, targeted and rapid distribution, is specificallyintended for policy and decision makers in the fields concerned.

The current Asian crisis ranks already among the most notable of themany crises in financial history, affecting many key emerging marketsbeyond the region. The resulting policy challenges have given urgency tothe debate on a coherent approach to financial globalisation. This PolicyBrief aims at informing that debate. It documents the extreme costs ofthe ongoing crisis and evaluates several suggestions for crisis resolutionand crisis prevention, both on a global scale and with respect to host-country policies.

3

After the Great Asian Slump: Towards a Coherent Approach to Global Capital Flows

Table of Contents

The Gains from Global Capital Mobility: Myth or Reality? ........................................... 6

The Damage Done by the Sudden Stop: The Asian Slump ........................................... 9

Crisis Management: Global Approaches ........................................................................... 12

How (Not) to Recover from the Domestic Debt Overhang ....................................... 17

Global Approaches to Crisis Prevention ........................................................................... 23

Crisis Prevention: What Host Countries Can Do.......................................................... 29

Notes .................................................................................................................................. 35

Bibliography .............................................................................................................................. 36

Table 1 Net Private Flows to Asian Countries in Crisis ........................................... 10

Table 2 Current Account Balance and Real GDP ....................................................... 11

Table 3 Exchange Rates, Stock Markets and Yield Spreadsin Asian Countries in Crisis ............................................................................... 11

Table 4 The Foreign Private Debt Overhang in Asian Countriesin Crisis, 1998 ....................................................................................................... 11

Table 5 The Domestic Private Debt Overhang in Asian Countriesin Crisis, 1998 ....................................................................................................... 12

Table 6 Possible Debtor-Creditor Relationships ........................................................ 15

Table 7 Cost of Debt, Credit Growth and Trade ...................................................... 18

Table 8 Bank Crisis Resolution in Asia, 1998 .............................................................. 19

Table 9 The Fiscal Impact of Banking and Currency Crises, 1990s ........................ 20

Table 10 Basle Capital Accord: Risk Weights by Selected Categoryof On-Balance-Sheet Assets .............................................................................. 25

Table 11 Maturity Structure of Foreign Debt in Selected Countries ....................... 34

Box 1. Banco Central de Chile Studies on the Effectiveness of the Country’sMeasures to Regulate Capital Flows ............................................................... 32

5


The 21st century is expected to see a growing share of global output movefrom the rapidly ageing OECD economies to the younger developing world(OECD, 1997). This will benefit both regions. With labour forces stagnating orshrinking in the OECD area and with pension assets increasingly decumulated,returns on capital will be depressed. All the world’s labour force growth will takeplace in the developing countries, promising higher returns on capital there(MacKellar and Reisen, 1998). This promise is reinforced by the observation thatpoor countries have a higher potential to grow than rich countries. Theseexpectations, though, would not materialise without substantial global capitalflows, efficiently allocated to their highest sustainable social rate of return. Thisproviso brings us back to the very end of the 20th century.

In 1998, the world has witnessed the strongest financial panic since the GreatDepression. Private capital flows, surging into the developing countries withunprecedented size in the years before, have made a sudden stop. What started asthe Thai baht crisis in July 1997, quickly spread to other East Asian countries — Indonesia, Korea, Malaysia and to some extent the Philippines. In these worst-hiteconomies, income and more so consumption levels have since fallen at an alarmingrate; the social costs (and political repercussions) are beginning to make themselvesfelt. By autumn 1998, crisis contagion had assumed global proportions, spreadingto virtually all emerging markets, as investors returned to core OECD safe-havenpaper. Before the “New Global Age” can materialise, therefore, this period ofhistoric policy challenges has given urgency to the debate on and the design of acoherent approach to financial globalisation.

This Policy Brief aims at informing that debate. First, it sets out the importanceof a coherent approach to globalising capital flows, and the potential benefits. Whilethese benefits have been at times oversold and certainly under-documented, areversal from financial globalisation would harm the industrial and developingcountries alike. Second, the Policy Brief documents the extreme economic cost ofthe ongoing financial crisis, and hence underlines the need for improved crisismanagement and crisis prevention. The following sections evaluate severalsuggestions to improve the management of a crisis, once it has erupted, both byglobal regulation and by the countries affected. The final sections are devoted tocrisis prevention, evaluating what can be done, both by global and by recipient-country regulation, to reduce the frequency and severity of financial and currencycrises.

6


The Gains from Global Capital Mobility: Myth or Reality?

With virtually all emerging-market assets on fire sale, these are very hardtimes to “sell” the gains from global capital mobility. While these gains do certainlyexist, the quantitative evidence is surprisingly sketchy. In principle, the benefits ofglobal capital mobility should apply particularly in the interaction between thecapital-rich, moderately-growing and fast-ageing OECD economies and the capital-poor, fast-growing and slowly-ageing emerging economies. The gains would resultfrom a better allocation of world savings to the most productive investmentopportunities, and the possibility of maintaining consumption levels in the event ofadverse shocks and of demographic trends. Moreover, it is often held that opencapital markets impose higher standards of economic policy on capital-recipientcountries. Even with net capital flows between the OECD and non-OECDeconomies balanced, open capital markets can be presumed to offer sizeablediversification benefits and spillovers in the form of technology, managerial know-how, market access and competition dynamics. Differences between the two areaswith respect to the exposure to country-specific shocks as well as the stage ofeconomic and demographic maturity suggest that the diversification benefits offinancial globalisation will not disappear quickly.

Three major currency and financial crises in the 1990s, however, in countrieswith broadly sound macroeconomic fundamentals (Scandinavia1991-93, Mexico 1994-95, now Asia), have alerted economists to the magnitude ofthe potential costs of open capital markets. In view of the deep economic, socialand political disruptions of these crises, some countries have started to retreattowards financial autarky by imposing capital outflow controls or by unilaterallydefaulting on their foreign liabilities. The events fully confirmed earlier warningsthat the macroeconomic adjustment to a sudden reversal of foreign capital flowscan be extremely painful. Governments and central banks had been advised (e.g., ina series of OECD Development Centre studies, originating with Fischer and Reisen,1992) to beware of the sustainability of capital inflows. In an address to Asianmonetary authorities in the autumn of 1995, the following reasons for cautionabout capital flow reversals were spelled out (Reisen, 1996, p. 73):

“First, it is increasingly acknowledged that global capital markets sufferfrom three major distortions: the problem of asymmetric informationcauses herd behaviour among investors and, in good times, congestionproblems; the fact that some market participants are too big to failcauses excessive risk taking. It is questionable, therefore, whether thefinancial markets will discipline governments into better policies; evenif they were to do so, the social and economic costs may be excessive.

7


Second, any shortfall in capital inflows will require immediate cutbacks indomestic absorption to restore external balance. The savings-investmentbalance is more likely to be achieved through cuts in investment thanthrough higher savings in the short term, compromising future outputlevels. Current output levels fall to the extent that rigidities preventresource reallocation, so that contractionary disabsorption effectsoutweigh expansionary substitution effects.

Third, the expansion of domestic credit connected with unsterilisedcapital inflows may not be sound enough to stand the rise in domesticinterest rates and the fall in domestic asset prices that go with a reversalof these inflows. The resulting breakdown of domestic financial institutionsprovides incentives for monetary expansion and fiscal deficits incurredby the public bail-out of ailing banks.”

The burden of proof of the gains from free capital flows has shifted to theproponents of open capital markets who are being criticised for having offeredmore “banner-waving” than hard quantitative evidence on the benefits of financialglobalisation (Bhagwati, 1998). A look at the numbers seems to suggest that mostof the gains that developing countries could reap from financial openness wereobtained by foreign direct investment inflows:

— First, if foreign savings permit an acceleration of investment by augmenting(rather than crowding out) domestic savings, they typically have a positivetemporary GDP growth effect of half a percentage point (Reisen, 1996). Thisassumes typical capital shares and capital output ratios as well as a net inflowof 3 to 4 per cent of GDP; furthermore, any externalities arising fromopenness are ignored.

— Second, foreign direct investment adds both to domestic investment and tolong-term growth if the host-county is largely undistorted1. Borensztein, DeGregorio and Lee (1995) find that for each percentage point increase in theFDI-GDP ratio, the rate of growth in the host economy increases by0.8 percentage points. The contribution to long-term growth results fromtwo effects. First, foreign direct investment adds to domestic investment, asboth are complementary in production and through positive spillover effects.Second, foreign direct investment stimulates growth through the embodiedtransfer of technology and efficiency, provided the host country has aminimum threshold stock of human capital.

8


Third, there is little evidence in the data that countries without capitalcontrols have grown faster than countries with capital controls, aftercontrolling for growth determinants such as income and educationlevels (Grilli and Milesi-Ferretti, 1995; Rodrik, 1998). These studies,however, do not allow for varying degrees of intensity of capital accountrestrictions, nor for the different growth impact of various capital-account items (Eichengreen, Mussa et al., 1998). Except for foreigndirect investment, the time series for private capital flows are not yetlong enough to draw strong policy conclusions. In particular, there is nocross-country study that would investigate the impact of capital accountliberalisation, while controlling for the strength of the domestic financialsystem. It has been noted that none of the developed OECD countriesmaintain general capital controls, even on short-term capital (Poret,1998). This may indicate that with mature financial systems, the benefitsof free capital mobility largely outweigh any costs. It may also indicatethat mature OECD economies are subject to less violent shocks ininvestor sentiment, and hence less disruption, than are developingcountries.

The virulence of the 1997-98 contagion also reflected new financial technologiesand highly leveraged assets, giving speculative currency attacks unprecedentedspeed and force (Summers, 1998; IMF, 1998a). Private cross-border flows havebecome volatile for several reasons on the supply side, notably:

— A growing proportion of long asset positions has become leveraged, mostnotably in the case of hedge funds; leverage (investing with borrowed fundsworth a multiple of own funds) implies abrupt portfolio changes when thebanks (who lend money to the hedge funds) make “margin calls” (call in thecredits as the price of the collateral drops below a specified level). OECD-based banks not only lent to hedge funds, but engaged increasingly inproprietary trading activities directed at exploiting short-term tradingopportunities in the emerging markets, shorting low-coupon currencies suchas the yen and taking long positions in high-coupon emerging-market paper.The technique, dubbed yen or dollar “carry trade”, relies on low interest ratesin OECD markets and stable emerging-market currencies (for detaileddescription, see IMF, 1998a).

— The increased availability and variety of financial derivatives in the worldfinancial centres facilitate the evasion of emerging-market prudential regulationand supervision as well as of taxes and capital controls; and they obscure themeaning of capital account data from standard balance of payments accounts

9


(Garber, 1998). The unwinding of the underlying positions hastens andintensifies speculative attacks on pegged currencies and exacerbatesweaknesses in emerging-market systems.

— Modern risk management systems — endorsed by and imposed by industrialcountry regulators — have become a prime source for the contagion effectsof a crisis (Folkerts-Landau and Garber, 1998; Reisen and von Maltzan, 1998).Risk control systems operating on the basis of international variance-covariance matrices of securities returns, imply that market volatility in onecountry will automatically generate an upward re-estimate of credit andmarket risk in a correlated country, triggering automatic margin calls andtightening of credit lines. Risk control systems, and in some industrialcountries prudential regulations, also require that institutional and otherinvestors hold only investment grade securities so that a downgrading of acountry’s credit rating leads to an immediate sell-off of the affected assets andto the closing of new funding.

The Damage Done by the Sudden Stop: The Asian Slump

The five countries most damaged by the Asian crisis — Indonesia, Korea,Malaysia, the Philippines and Thailand — received net private capital inflows worth6.6 per cent of their combined GDP over the period 1995-96. The excessiveoptimism among international investors at that time was reflected in very low yieldspreads on their debt instruments (less than 100 basis points over Eurobond yields).In the second half of 1997, there was a sudden stop. The reversal of net flows from1996 to 1997 constituted a swing of 11 per cent of their combined GDP. Thebiggest swing came from commercial banks who had extended loans well into 1997,despite earlier warnings on overexposure from the Bank for International Settlements(BIS) and the Institute of International Finance (IIF). There was also an importantreversal of net portfolio investment. The only capital-account component provingits staying power — just as during Mexico’s 1994-95 crisis — was foreign directinvestment (Table 1).

A sudden stop in capital inflows must be met by a reduction in aggregatedemand. Indeed, if depleted foreign exchange reserves have to be rebuilt,disabsorption (the cut in consumption and investment) must even exceed thereversal in flows. Between 1996 and 1998, the required switch on the currentaccount of the five Asian countries in crisis, i.e. the difference between aggregatedemand and output, was nothing less than 14.5 per cent of their GDP. The size and

1 0


rapidity of the required adjustment has triggered a major economic growth crisisin the affected countries, exacerbated by weak banking systems. Used to growthrates in the 6-10 per cent range, the five Asian countries in crisis are forecast onaverage to shrink by 9 per cent in 1998 and a further 4 per cent in 1999 (Table 2).Official unemployment, traditionally lower than in OECD countries as safety nets(and claims) are absent, is expected to jump from low single-digit levels to 15 percent in Indonesia, 13 per cent in the Philippines and 9 per cent in Thailand,according to ILO estimates (ILO, 1998).

Table 1. Net Private Flows to Asian Countries in Crisis1

($ billion)

1995 1996 1997e 1998f 1999f

Private Net Flows 83.8 93.8 -6.0 -24.6 -15.1Commercial banks 58.0 58.3 -29.0 -30.5 -17.8Other debt (bonds) 9.9 18.1 23.3 -2.1 -3.8Portfolio equity 11.0 11.6 -6.8 1.1 -0.9Direct foreign investment 4.9 5.8 6.5 6.9 7.4

e = estimate, f = forecast1. Indonesia, Korea, Malaysia, Thailand, and the Philippines.Source: Institute of International Finance, Capital Flows to Emerging Market Economies, Washington, D.C., 29 September 1998.

The sharp withdrawal of private flows was reflected in tumbling exchangerates and falling local stock market prices as well as in the sovereign risk yieldspreads’ rising to nearly prohibitive levels. The Thai baht and the Korean won losthalf of their value against the dollar, the Indonesia rupiah 80 per cent in the firstmonths of the crisis, fanning a strong rise in non-performing loans in the localbanking system and wiping out net capital for unhedged corporate borrowers. Thereactive approach of the sovereign rating industry (Reisen and von Maltzan, 1999)intensified panic by downgrading Asian borrowers from “investment grade” to“junk” status. The lower country ratings forced institutional investors to offloadAsian assets and allowed banks to call in loans.

While there was initially a tendency to blame policy and institutionalshortcomings in the Asian crisis countries for the reversal in capital flows, thewidespread contagion of the crisis to other emerging markets judged fundamentallysound has strengthened the analysis of those economists who had argued that thereversal in flows, exchange rates and sovereign ratings in such a short period cannotbe attributed to changes in the affected countries’ fundamentals. Regardless of theultimate causes for the Asian slump, it has worsened the economic outlook of thecountries in crisis for years to come. Not least as a result of tumbling exchange ratesand strong GDP contraction, the crisis countries now face a private-sector

1 1


Table 2. Current Account Balance and Real GDP

1995 1996 1997e 1998f 1999f

1. Current Account (per cent of GDP)Asian Countries in Crisis1 -4.1 -5.1 -2.6 9.4 8.5Emerging Market Economies2 -2.0 -1.8 -1.4 -0.9 -0.5

2. Real GDP (per cent change)Asian Countries in Crisis1 8.4 7.0 4.5 -9.1 -3.8Emerging Market Economies2 4.4 5.0 5.0 1.2 1.4

e = estimate, f = forecast1. Indonesia, Korea, Malaysia, Thailand, and the Philippines.2. 29 Major Emerging Market Economies.Source: Institute of International Finance, Capital Flows to Emerging Market Economies, Washington, D.C., 29 September 1998.

Table 3. Exchange Rates, Stock Markets and Yield Spreads in Asian Countries in Crisis(30 September 1998)

Change since July 1997, percentage Spreads on BenchmarkEurobonds

Dollar Exchange Rate Stock Market Index (basis points)

Indonesia -77.4 -62.1 1 397Korea -39.3 -59.8 810Malaysia -33.7 -69.5 n.a.Thailand -34.0 -55.0 679Philippines -39.7 -55.3 821

Source: Reuters Online; Deutsche Bank Research, Global Emerging Markets, Vol. 1.3, October 1998.

Table 4. The Foreign Private Debt Overhang1 in Asian Countries in Crisis, 1998(30 September 1998)

Foreign Debt($ billion)

Discount(percentage)

Debt Overhang($ billion)

% GDP

Indonesia 79 55 44 40.0Korea 83 36 30 9.6Malaysia 39 36 14 18.7Thailand 79 35 28 17.5Philippines 24 41 10 15.4

1. The discount is calculated as the change in the market price of the debt since 30 June 1997. The debt refers toestimated market-based debt (total minus estimated official). The debt overhang is estimated by applying thediscount to all commercial external debt.

Source : Deutsche Bank Research, Global Emerging Markets, Vol. 1.3, October 1998.

1 2


debt overhang that is estimated to exceed the 1980s Latin American proportionsby far (Armstrong and Spencer, 1998). The resolution of this private debt overhangwill inevitably burden government budgets, because it necessitates a large build-upin public sector debt, and it will require foreign investors to write down some oftheir claims. Non-performing loans, except in the Philippines, exceed bank capitalby far, and discounts on external debt signal that at least some of the net presentvalue of the debt is not expected to be repaid. Any delays in bank restructuring willraise the share of non-performing loans; any delays in foreign debt relief will leadto higher secondary market discounts; such delays can only credit-starve profitableactivities and postpone the return of investor confidence longer than necessary.For the time being, the unprecedented withdrawal of foreign capital from Asiaconfronts the Asian countries with a transfer problem, just as it did over the 1980sin Latin America. The lesson was then (Reisen and van Trotsenburg, 1988) that thebudgetary problem was “solved” by high inflation as regular tax receipts wereinsufficient to pay for the transfer and that declining export prices constituted asecondary burden as a result of many countries simultaneously trying to producetrade surpluses.

Crisis Management: Global Approaches

Thorny policy issues arise for global crisis resolution that are difficult tobalance. On the one hand, a generalised lack of confidence can only be reversedthrough an effective lender of last resort who could credibly commit sufficientliquidity in support of any country deemed fundamentally sound but illiquid. On theother hand, a successful rescue operation bails out (at least partly) investors whoshould have taken losses on their excessive risk exposure and thus is apt to

Table 5. The Domestic Private Debt Overhang in Asian Countries in Crisis, 1998

Non-performing loans Loan losses% of loans % of GDP $ billion % of bank capital

Indonesia 70 19.0 20.8 1 088Korea 35 17.2 53.0 196Malaysia 35 20.3 15.1 145Thailand 45 26.5 41.6 347Philippines 20 6.5 4.1 86

Source: Deutsche Bank Research, Global Emerging Markets, Vol. 1.3, October 1998.

1 3


encourage future excessive risk taking, in other words: moral hazard. The moralhazard argument, though, is easily exaggerated. The evidence from the Asian crisisindicates, as indeed does the evidence from Mexico’s 1994-95 crisis, that equity andbond holders have experienced heavy losses. Commercial and investment banks,by contrast, can be perceived as having been bailed out during previous crises andhaving suffered only limited losses in the Asian crisis so far. While such perceptionholds for the banks’ balance sheet exposure, the level of losses taken from off-balance-sheet exposures and activities (such as securities underwriting) remains tobe seen.

It is now understood that the international financial institutions cannot belenders of last resort in a world of intense capital mobility. In earlier quiet decadeswhen capital mobility was limited, balance-of-payments crises meant imbalance oncurrent account. Then, the phased conditional support from these institutions’resources was effective in helping countries towards payments balance. Today,with balance of payments, exchange rates and interest rates governed by the capitalaccount, the international financial institutions need to frontload their assistancemassively to have a market impact, while their involvement may feed further doubtsamong private investors. Official resources are, and will remain, insufficient in aworld where private liquid assets are worth trillions of dollars and are virtually freeto move across borders. Moreover, derivatives and leveraged positions havemultiplied the market impact that the moving of these assets can exert. To theextent, therefore, that the world’s leading central banks are not prepared toassume the task of a lender of last resort, any emerging-market investment will haveto reckon more downside risk than they would have had to just a few years ago.This may partly reverse financial globalisation and reduce private flows to developingcountries. It may, however, raise the quality of these flows as bail-out induced moralhazard is reduced.

The limited resources of the international financial institutions can be givensome leverage, however. In October 1998, the first successful issue of a Thai bondsince the crisis erupted was made possible, despite the country’s junk credit-ratingstatus, because of a World Bank guarantee that lifted the bond to investment grade.By providing guarantees instead of money, the international financial institutionscan use their funds more productively and help selected countries to re-entercapital markets. In order to reduce the risk of default on these bonds, only countrieswith considerable reform progress should be considered; the guarantee should alsobe tied to specific projects rather than balance-of-payments finance. This latterproviso clearly, however, limits their scope in the resolution of the current crisis.

1 4


The G22 Working Group on International Financial Crises, which publishedits recommendations for managing future crises in October 1998 (stating explicitlythat the recommendations should not be considered as an agenda for theresolution of the 1998 crisis), acknowledges that “...the scale of private capital flowssignificantly exceeds the resources that can reasonably be provided by the officialcommunity”. (Summary of Reports, p. 21). With official resources insufficient, thisimplies that the Asian foreign debt overhang documented in Table 4 will not beresolved without some debt relief. While the G22 report favours voluntary debtreduction, it concedes that “...a purely voluntary approach may be impractical. Inparticular, it might consume so much time that it would lead to an erosion ofconfidence that would be contrary to the collective interest of creditors anddebtors in a co-operative and equitable workout” (Summary of Reports, p. 22).Indeed, this is precisely what the 1980s debt crisis has taught us and what the risingyield spreads on Asian sovereign bonds are telling us now.

The 1980s Latin American debt crisis showed that debt rescheduling aloneprovides little relief in the presence of insolvency or severe illiquidity problems.This was finally recognised by the 1989 Brady debt reduction initiative aftersecondary market discounts on Latin American liabilities had continuously risen toan average of 70 per cent. Likewise, while during the early months of the currentAsian crisis fairly stable yield spreads reflected the market perception that the crisiswas a temporary liquidity problem, the growing awareness of insolvency and defaultrisk fuelled a rise in secondary market discounts to 40 per cent, on average, for thefive Asian countries in crisis (see Table 4). As emphasised by Armstrong andSpencer (1998) of Deutsche Bank Research, there has been very little external debtreduction in Asia so far, as interbank debt rescheduling arrangements resulted inno reduction of the debt stocks and as the Indonesia plan for corporate debt hadno voluntary contributors.

Co-ordination failures essentially prevent voluntary debt reductions, as thesefail to address the “free-rider problem”. By voluntarily writing off a part of its claim,a creditor produces an externality to all other creditors because the price of theremaining outstanding debt will rise. Since the one making the voluntary write-offcannot “internalise” the rise in securities prices, and thus it favours all othercreditors (and investors), too little of the “public good” of debt relief will occur ona voluntary basis. Debt reduction, in order to be effective in stimulating Asia’srecovery, needs a centrally co-ordinated approach. So far, this is not even on thepolicy agenda. Leadership is required to avoid the protracted round of negotiationsbetween debtors, creditors and their respective governments that proved sodevastating in Latin America during the 1980s.

1 5


Table 6 explains why co-ordinated debt reduction will be very difficult toorganise. The 1980s was characterised by one debtor — the public sector — anda fairly homogenous group of syndicated bank creditors (Case No. 1). Internationalbanks used the London Club as a forum for rescheduling and reducing public sectordebt. Case No. 2 largely reflects Mexico in 1995 where the public-sector debtorwas confronted with a heterogeneous group of bondholders. The co-ordinationproblem was overcome by a then unprecedented IMF loan and by swift US support,leading to Mexico’s quick recovery. Mexico’s bail out, however, has been heldresponsible by some for encouraging moral hazard by international investors insubsequent years and, ultimately, the current global financial crisis (e.g. Regling,1998). Cases 3, 5 and 6 represent the current situation for the Asian countries incrisis, where essentially private-sector debtors and creditors are involved. With alarge number of debtors and creditors on both sides, it is hard to start negotiationsand even harder to find agreement among the participants.

It is encouraging that Group of Ten (1996) recommendations for theresolution of sovereign liquidity crises, spelled out in the aftermath of Mexico’s1994-95 crisis, but not followed by co-ordinated policy actions since then, are nowbeing actively pursued in order to bail foreign creditors in, rather than out (Groupof Twenty Two, 1998):

— Debt holders are to be induced to participate actively in the resolution of debtcrises through debt contracts that would i) provide for the collectiverepresentation of debt holders in the event of crises; ii) allow for qualifiedmajority voting to alter the terms and conditions of debt contracts; andiii) require the sharing of assets received from the debtor.

Table 6. Possible Debtor-Creditor Relationships

Debtor Creditor

Private Banks Private non-banks(Bond markets)

Public Sector Case No. 1:London Club

Case No. 2:Mexico 1995

Private Banks Case No. 3:Korea, Thailand

Case No. 4:–

Private non-banks Case No. 5:Essentially Indonesia

Case No. 6:Partially Indonesia

Note: Creditor-debtor relationships shown for individual countries represent the situation at the start of the crisis.Source: Regling (1998).

1 6


— The IMF Executive Board has agreed that the Fund should extend the 1989decision on “lending into arrears” in order to provide support for debtorcountries with sovereign arrears to private creditors (including bond holders),and members with non-sovereign arrears to private creditors. The “lendinginto arrears” could signal confidence in the debtor countries’ policies andprospects as well as hasten creditors in concluding debt rescheduling orreduction deals.

These provisions, however, should be more effective in reducing the frequencyand severity of future crises than in resolving the current emerging-market crisis.

If unilateral debt moratoria are to be avoided, a partial socialisation ofdistressed foreign private sector debt by the Asian crisis countries in return forrespective concessions by foreign creditors in terms of pricing, maturity andprincipal of their claims has to be envisaged. Debt socialisation results from grantingof government guarantees (as in Korea late 1997) on private sector debt, its legalassumption or its conversion into public debt. This moves the Asian borrowersback to Cases No. 1 and 2 in Regling’s table.

The quid pro quo must come from foreign lenders. They, however, can affordto wait (with sizeable provisions already made in most cases), in the hope thatforeign aid flows and funding from international financial institutions will raise themarket value of their claims. Creditor country governments are therefore welladvised to induce the private sector, and their home banks in particular, intofinancial rescue operations, not only to reduce moral hazard in future private sectorlending but also to encourage Asia’s recovery now. A model of how the complexunderlying co-ordination and negotiation problems can be solved, was provided bythe Brady plan (Reisen, 1994; Dooley, 1994; Cline, 1995). The Brady plan becamepossible only after debtor and creditor governments forced creditor banks toparticipate in the deal; in turn, a broad menu of options to creditors, allowing theresulting deals to combine debt reduction, rescheduling and the provision of newmoney, raised the attractiveness of the plan to creditor participation. The Bradyplan also showed that only a part of the debt overhang had to be written off to makethe remaining debt sustainable, as the post-relief market price of debt rose.

1 7


How (Not) to Recover from the Domestic Debt Overhang

The domestic debt overhang (see Table 5) — primarily bank obligationswhich can no longer be serviced — in the Asian countries in crisis is larger than boththeir external debt overhang (except in Indonesia and the Philippines) and than thatwitnessed in prior financial and currency crises in Latin America and Scandinavia.Whatever the external financial and trade conditions, the resolution of Asia’sdomestic debt problem must be the overriding policy concern for Asian governments.The write-down of claims on insolvent corporate borrowers crystalises the losseson the banks’ balance sheets. Without resolving the non-performing loan problemand recapitalising the domestic financial systems, there will be no sustainedrecovery. But without economic recovery, Asia’s financial systems and corporatesectors will be further impaired, as fiscal imbalances ultimately needed for bankcrisis resolution would become unsustainable.

Banking and currency woes are costly. A recent study by the IMF (WEO,1998), covering 50 crisis episodes over the 1975-97 period, indicates that thecumulative loss in output is severe — on average some 14-15 per cent of GDP.Average recovery came sooner in emerging market economies (2.8 years) than inindustrial countries (4.1 years), though the cumulative output loss was larger, onaverage, for emerging market economies than for industrial countries. Suchhistorical evidence, in view of the size of the domestic debt overhang in the Asiancrisis economies, warns against being too relaxed about Asia’s potential to growout of its problems. Basing policy responses on overly optimistic scenarios makesit difficult to develop policies against worse outcomes, thus exacerbating thedownturn.

A case in point is Japan’s experience over the 1990s. It was hoped that bankswould “grow out of” their problems with a magic mix of wider intermediationspreads (imposing rescue costs on borrowers and depositors), relaxation ofprudential regulations and lower funding rates. But with low levels of capitalisation,Japanese banks proved unwilling to lend; with declining collateral values, goodcredit risk evaporated. The failure to address the consolidation of the domesticfinancial system led to a decade of stagnation (Posen, 1998).

1 8


The recent stabilisation of currencies, the recent drop in money market ratesand the dramatic turnaround in trade balances in the Asian crisis countries can be,and indeed often has been, interpreted as a first solid sign of “bottoming out”; theyhave also underpinned optimistic growth projections already for the closingmonths of 1998. However, the Asian currencies have been supported by officialflows (with relatively little more to follow); moreover, having undershot equilibriumlevels so drastically at the start of the crisis (Davies, 1998), a gradual realappreciation was widely expected to materialise. Relying on a gradual drop inmoney market rates to stimulate growth ignores the economics of banking crises.While the drop in money rates may just reflect rising demand for risk-less cashbalances, the cost of capital is still higher than domestic private-sector creditgrowth (except for Indonesia), implying tight rather than easy credit conditions.Finally, the trade adjustment has been achieved through import compression, asexport growth has been flat or falling. Unless trade is redirected away from intra-regional orientation (roughly 50 per cent now), unless exporters cease to becredit-starved and unless Asia recovers there will not be much export-led growth(Table 7).

Tumbling exchange rates and high interest rates have driven corporatebalance-sheet losses beyond equity values in a large number of firms in Indonesia,Korea and Thailand. According to this criterion, the World Bank (1998) assessestwo out of three listed firms as bankrupt in Indonesia, two out of five in Korea, andone out of four in Thailand. Even very loose fiscal and monetary policies will fail tostimulate a recovery, unless restructuring and debt workouts are carried outupfront. The domestic debt overhang is mirrored in a systemic banking crisis as,with rising default, banks become unwilling to provide new loans and prefer toaccumulate risk-less assets in order to regain capital adequacy standards.

Table 7. Cost of Debt, Credit Growth and Tradeannual percentage change

Cost of Debt1, % Private Dom. Credit2 Exports, $ Imports, $1997Q2 1998Q2 1997Q2 1998Q2 1997e 1998f 1997e 1998f

Indonesia 15.4 41.0 25.4 104.1 12.2 -1.3 4.5 -31.6Korea 9.6 18.6 21.3 8.7 6.6 -7.1 -2.1 -36.2Malaysia 7.8 12.3 n.a. 10.3 0.9 -9.3 1.5 -15.3Thailand 20.4 23.1 16.2 15.2 4.2 -0.4 -13.8 -28.7Philippines 12.6 22.6 33.1 12.6 22.9 15.9 14.1 -9.9

e = estimate; f = forecast1. Six months local interbank rates (Korea, six months implied offshore rates) and corporate spread over local

interbank rates.2. End of period.Sources: Deutsche Bank Research, Global Emerging Markets, Vol. 1.3, October 1998; JP Morgan, World Financial Markets, Fourth

Quarter 1998.

1 9


As a large share of the financial and corporate sectors have been renderedtechnically insolvent by the new exchange rate/debt cost combination, governmentshave to assume a strong leadership role in revitalising the private sector. The OECD(1998) and the World Bank (1998) draw the following core lessons from successfulrestructuring experiences:

— Early and comprehensive evaluation (distinguishing between viable andbankrupt banks);

— Transfer of non-performing loans from the banks’ balance sheets to a separateloan recovery agency;

— Provision of capital only to viable banks during restructuring;

— Enforcement of exit policies for firms (bankruptcy codes, asset sales);

— Requirement for loan workouts to recover part of the fiscal cost ofrestructuring and to send signals to delinquent borrowers.

Beyond this list, the attraction of foreign banking capital is a further option torecapitalise banks (and improve banking standards). Before the crisis, foreign bankshad relatively little penetration of Asia’s domestic markets. The IMF (1998a) warns,however, that this may further weaken domestic banks (which stay with the“lemons”, i.e. bad credit risk). While the Asian crisis countries now have mostlylifted bank entry restrictions, only Thailand has so far attracted new capital (worthroughly 5 per cent of GDP) into the banking system.

Table 8. Bank Crisis Resolution in Asia, 1998

Indonesia Korea Malaysia Thailand

Number of institutionsOriginal 240 169 89 132Closed 23 21 0 56Nationalised 47 2 0 13Merged 0 5 61 13Bought by foreigners 0 0 0 4

Source: JP Morgan, World Financial Markets, Fourth Quarter 1998.

Market observers agree that Korea and Thailand have progressed more onbanking sector reform than Malaysia and Indonesia (the Philippines, as shown, doesnot have a banking crisis as the preceding credit boom started relatively late there).As shown in Table 8, the weakest banks have been closed down and been liquidatedin auctions, some banks have been nationalised with the intention of later

2 0


privatisation, and mergers have been forced. Moreover, restructuring agencies andbankruptcy laws have been established, and stricter loan provisioning, assetclassification and income recognition requirements have been introduced. Despitesuch progress, the fear of causing further bankruptcies in the corporate sector andthe implicit fiscal cost have limited and delayed the solution of banking and non-performing loan problems (see Table 9 below).

Table 9. The Fiscal Impact of Banking and Currency Crises, 1990s

Years Non-performing loanspercentage of total loans at

peak

Fiscal and quasi-fiscal cost1

percentage of GDPPublic revenues

percentage of GDP

Finland 1991-93 9 8-10 33Norway 1991-93 9 4 40Sweden 1991-93 11 4-5 39

Mexico 1994-95 11 12-15 15

Indonesia 1998 70 17 11Korea 1998 35 16 20Malaysia 1998 35 15 23Thailand 1998 45 18 20

1. Lower estimates include costs of funds, credit, and bonds injected directly into the banking system; higher estimatesinclude other fiscal costs, such as exchange rate subsidies.

Sources: IMF, World Economic Outlook 1998; IMF, International Financial Statistics, various issues; Deutsche Bank Research, GlobalEmerging Markets, Vol. 1.3, October 1998.

Delay in bank restructuring to ease the fiscal burden involved and to allowcontinuing expansion of bank credit to unviable borrowers will only raise theultimate costs of financial system reform; this lesson has been brought home clearlyby Latin America’s experiences in the 1980s (World Bank, 1998) and Japan’s in the1990s (Posen, 1998). While for speedy recovery it is primordial to restore creditto exporters and to viable investment, delay of reform will provide perverseincentives to under-capitalised or bankrupt banks. Either, to stem the decline inrisk-weighted capital ratios, banks will increase their exposure to governmentliabilities and other zero-risk weighted assets. Or, banks will engage in activities(such as derivatives) with a high risk-return profile in a gamble to earn their way outof difficulties. Good risks, by contrast, are underfinanced and growth prospectsundermined as long as the banking crisis is not fully and quickly addressed.

2 1


There is a clear trade-off, however, between cleaning up the banking systemaggressively to eliminate a major obstacle to growth, and weakening the government’sfiscal position. This trade-off has never been stronger than it is for the Asiancountries in crisis. Not only are the loan losses in the Asian crisis countries muchhigher than in earlier episodes of banking and currency crises, but they also implyfiscal costs that must be met from much lower public revenue levels than everbefore (Table 9). The fiscal costs stem mostly from the need for the governmentto take over non-performing assets by issuing a corresponding amount of new debt.In the Asian crisis countries, the interest costs alone on the new government debtrepresent a very large share of government revenues; further fiscal burdens arisefrom direct equity stakes.

Given these strong policy dilemmas, there are some tempting but short-sighted and ineffective policy prescriptions for reducing the fiscal cost of cleaningup the debt overhang. Notably, the imposition of controls on capital outflows orunilateral debt default clearly damage other emerging markets.

On 1st September 1998, Malaysia imposed strong controls on outflows. Asa crisis measure, its main rationale is to lower the fiscal cost of bank restructuringby lowering the local interest rate without leading to further currency depreciationwhich would in turn nurture more losses in the banking system and in unhedgedcorporate balance sheets. The developing-country evidence, however, is not goodfor outflow controls (Dooley, 1995). They generally help delay reform, allowextending further credit to favoured borrowers and hence are akin to deepen thenon-performing loan problem in the local banking system, thus raising the ultimatecost of bank restructuring. There is a danger that Malaysia will closely follow thisscript. Outflow controls effectively discourage inflows, including foreign directinvestment that has helped Malaysia to grow in the past. The absence of inflows inturn lowers the equilibrium exchange rate, and with the nominal exchange ratefixed by decree, raises the black market premium. A rising premium intensifiesincentives to evade the outflow controls through overinvoicing of imports orunderinvoicing of exports or through stimulating purchases of real commoditiessuch as gold.

Two weeks prior to the Malaysian announcement Russia not only imposedoutflow controls but also defaulted by announcing a 90-day moratorium on itsprivate obligations (involving both foreign and domestic creditors). Latin America’s1980s experiences with (often temporary) default have been extremely costly and

2 2


thus short-lived (Reisen, 1989). The costs include: loss of market access reflectedin skyrocketing sovereign yield spreads; the transfer of official reserves to avoidseizure; support for foreign affiliates of domestic banks that are being precludedfrom interbank business; private capital flight resulting from lost confidence ofdomestic and foreign investors; and, last but not least, the cut off from trade-relatedcredit lines. Moreover, as domestic banks are often important (captive) lenders totheir government and as default dries up market liquidity, banks will see their capitalfurther depleted.

Other policies to reduce the fiscal cost of bank crisis resolution also haveserious side effects (see Reisen, 1989). Chile, for example, recapitalised the bankingsystem in the early 1980s by issuing Central Bank debt rather than governmentdebt. While this leaves public budgets unaffected by the crisis resolution, the quasi-fiscal costs are high. Ultimately, the government had to recapitalise the CentralBank; but, admittedly, breathing space was gained in the meantime. A moretransparent way is to follow the Scandinavian practice of establishing independentloan recovery agencies. These were mandated to maximise the recovery value ofnon-performing assets by either immediate closure and sale of the underlyingcollateral or by trying to recapitalise illiquid borrowers to improve repaymentprospects.

Another option is to tax government bond returns. This raises the tax baseby the amount of public interest outlays on government debt; but, if there is perfectforesight and if assets are perfect substitutes, taxing interest payments has no effecton government budgets. Changes in tax rates on any assets bring about an equalchange in their equilibrium returns, and hence leave after-tax yields unaltered.

Yet another policy option to be explored is the concept of debt/equity swaps,which were an important element to the solution of Chile’s debt overhang (bothdomestic and foreign) in the early 1980s. A debt/equity swap improves thestructure of corporate liabilities by reducing current debt service requirementswhile allowing creditors (the bank or the government, after the transfer of badloans) to reap the benefits of foregoing current debt service claims by sharing infuture corporate profits. However, swaps are difficult to implement on a large scale,in view of problems in loan valuation, equity pricing and corresponding free-riderproblems similar to those discussed above.

More importantly, the inevitable fiscal cost in Asia’s restructuring processmust — and can — be supported by growth-oriented fiscal adjustment and easymonetary policy. With extremely high domestic interest rates, banks will not startlending again; they rather invest money in high-interest public paper. The private

2 3


corporate sector, obviously, can not operate at interest rates that are 30-50 percent in inflation-adjusted terms. With negative or zero real growth, issuinggovernment debt at these rates will lead to explosive debt dynamics and governmentinsolvency — with default or hyperinflation as ultimate policy alternatives.

The fact that exchange rates have undershot their equilibrium levels by far canbe exploited by monetary policy. As shown by Ize and Ortiz (1987) for Mexico’s1980s debt overhang, real interest rates on domestic debt can fall, in an openeconomy, provided that the exchange rate initially overshoots in reaction to a debtand currency crisis. If the initial depreciation gives rise to expectations of futureappreciation, hence creating a wedge between returns in domestic and foreigncurrencies, this would allow debt servicing on local currency debt to fall. A moreaccommodating monetary policy can thus be part of continued restructuring incorporate and banking sectors, without compromising exchange rate stability, asconfidence is restored.

Fiscal adjustment to cope with the banking woes should not be sought at thecost of output growth, as it is very likely to be disrupted by social and politicalresistance. Fiscal adjustment also has to contain tax-induced capital flight. Thisimplies that the focus in Asian public finance has to be on tax-base broadeningrather than on non-interest spending cuts. Effective tax ratios are low in the Asiandebtor countries, and there are non-distortionary ways to increase them. Whilekeeping marginal tax rates low to discourage capital flight and encourage business,tax bases need broadening by eliminating exemptions and special incentives. Lowtax rates should raise compliance and enforcement of taxation, helped by highcredible penalties on outright avoidance and abolition of discretionary elements intax legislation. The tax base can also be broadened and the public revenue ratiobe raised, through introducing effective withholding schemes on wages, dividendsand interest and through strengthening tax administration to cross-checkdifferent tax sources.

Global Approaches to Crisis Prevention

The global 1997-98 financial crisis has presented complex and unprecedentedregulatory challenges to the international policy community. Attention hasincreasingly shifted from policy failures in emerging-market economies to thestructural weaknesses of the international financial system, “from the follies ofborrowers to the follies of lenders” (Laura D’Andrea Tyson, 1998). US DeputyTreasury L.H. Summers (1998) has described the self-fulfilling flight of private

2 4


capital out of the emerging-market economies as analogous to bank runs on entireeconomies. The G22 Report of the Working Group on International FinancialCrises has aimed at identifying policies to “prevent” international financial crises(and to facilitate their orderly resolution).

History suggests (Kindleberger, 1978) that financial and currency crises willnever be prevented altogether. Whatever regulatory changes the authorities mightfind the vision, determination and leadership to bring forward, these changes canat best reduce the frequency, severity and contagion of financial crises; they cannotpreclude them altogether. In the absence of 19th century “gunboat” diplomacy andin the absence of enforceable global bankruptcy legislation, capital flows betweenprivate or public entities of sovereign nations will remain prone to moral hazard,adverse selection and changes in investor sentiment. Progress, however, towardsa less crisis-prone international financial system has been made since late 1997, asofficial thinking has rapidly evolved.

Some proposals for a “new global financial architecture” have widespreadagreement among policymakers in both industrial and developing countries. Theneed for more transparency and accountability is generally accepted. The G22Working Group on Transparency and Accountability has attached particularimportance to enhancing the relevance, reliability, comparability and understandabilityof information disclosed by the private sector; the Group sees also a need forbroad, frequent and timely data on international exposures of investment banks,hedge funds and other institutional investors. The public sectors need to providebetter information disclosure on foreign exchange reserves, external debt andfinancial-sector soundness. An increasing number of countries have committedthemselves to the IMF’s Special Data Dissemination Standard, and BIS recording ofthe maturity, sectoral and national distribution of interbank lending comes in withbroader country and risk coverage and quarterly frequency.

Good accounting standards and complete, accurate and timely informationdisclosure are not only a necessary precondition for prudential regulation andsupervision. They also can help stabilise market expectations. To be sure, betterinformation is a necessary, not a sufficient condition, to prevent crises. “The Asianexperience makes this very clear: In spite of the ready availability of BIS datashowing the increasing vulnerability of some of these countries to a suddenwithdrawal of short-term international bank loans, the volume of these loans simplykept on rising”, notes the BIS in its 1998 Annual Report. To stabilise marketexpectations, points out Alice Rivlin (1998) “...people need to actually want to look— and too often those who are making profits would rather not hear bad news”.

2 5


It is also widely agreed that cross-border bank lending faces regulatorydistortions through the 1988 Basle Accord, the capital adequacy regime imposingdifferent risk weights by category of bank lending. Table 10 provides a selectiveoverview of the current risk-weighting scheme for on-balance-sheet assets. Mostimportantly, short-term bank credit to non-OECD banks of up to one year carriesa low 20 per cent risk weight, while long-term credit to non-OECD banks (overone year) is discouraged by a 100 per cent risk weight. A lower risk weight reducesborrowing costs, as banks have to acquire less capital relative to their risk-weightedassets. Similar distortions are created by the fact that claims on banks carry a 20 percent risk weight, while claims on the private sector carry a 100 per cent risk weight.This encourages cross-border interbank lending, which has been described as the“Achilles’ heel” of the international financial system (Greenspan, 1998). It furtherimplies that there is a greater incentive for banks to lend to unregulated hedge funds(indirectly through interbank lending) than to even the bluest of the world’s blue-chip companies. Note also that OECD-based banks and governments receive amore lenient treatment in the Basle Accord, even if they constitute sovereign risksequivalent or inferior to non-OECD emerging markets. Finally, the fixed 8 per centminimum capital assigned to risk-weighted assets works in a pro-cyclical way: At thepeak of the cycle, when asset prices are up, the capital buffer may be insufficient inlight of the higher downside price risk of collateralised assets; at the trough of acycle, by contrast, the Basle Accord may well intensify credit starvation.

Table 10. Basle Capital Accord: Risk Weights by Selected Categoryof On-Balance-Sheet Assets

Risk Weight Category

0 per cent • Claims on central governments and central banks denominated and funded in nationalcurrency

• Other claims on OECD central governments and central banks

20 per cent • Claims on multilateral development banks• Claims on banks incorporated in the OECD• Claims on banks outside the OECD with a residual maturity of up to one year

100 per cent • Claims on banks outside the OECD with a residual maturity of over one year• Claims on the private sector• Claims on governments outside the OECD, unless denominated in national currency

Source: IMF, International Capital Markets, September 1998.

2 6


The Asian crisis has also demonstrated that the recent shift in bank supervisionfrom rules-based to risk-focused methods is unlikely to tame volatile bank lendingbehaviour. The growing competition from other institutional investors, theglobalisation of banking, new technologies (such as the Internet) and the proliferationof new financial instruments have lowered the potential franchise value of banks andpushed them into assuming new market, credit and liquidity risks (BIS, 1998). Theregulatory framework, illustrated by the recent amendment to the Basle CapitalAccord, has tried to catch up with the increasing complexity of financial marketsby permitting the use of internal models for evaluating market risk. The state of theart risk management methodology — the Value at Risk (VAR) approach, anoutgrowth of portfolio theory — has been shown to have first encouragedexcessive bank lending and then intensified the global contagion of crisis. Banklending to emerging markets was overly encouraged by VAR models, whichestablish backward looking variance-covariance matrices on daily asset returns, andhence failed to signal the true extent of losses that certain long or short assetpositions could inflict (not surprisingly, the same model was used at the near-bankrupt Long Term Capital Management hedge fund). Contagion was intensifiedas a volatility event in one country automatically generated an upward re-estimateof credit and market risk in a correlated country. This triggered automatic margincalls and tightening of credit lines, as the VAR method encourages a defined netasset position (through, e.g., limiting net exposure to emerging markets by goinglong and short on highly correlated currencies, bonds or stock markets). Theapplication of internal risk control methods may also explain why Russia’s defaultand Malaysia’s controls had such a deep impact on other emerging markets(Folkerts-Landau and Garber, 1998).

Banks have also increasingly engaged in proprietary trading activities directedat exploiting short-term trading opportunities in the emerging markets; moreover,they have increasingly lent to unregulated hedge funds2 whose lack of transparencymasks the banks’ ultimate exposure to risk. The winding and unwinding of theunderlying positions, often leveraged to high multiples in order to enhance returns,has strongly destabilising effects in the shallow asset markets of developingcountries. This not only generates boom-bust cycles in the flow-recipient countries;it also raises counterparty risk so that banks and hedge funds can no longer beclosed rapidly by regulators without posing a wider economic threat. It follows thatthe Basle capital adequacy requirements would have to raise the cost of proprietarytrading and of bank lending to hedge funds. It has to be reckoned, however, thatthe sprawling derivatives facilitate the evasion of prudential regulation and supervision,

2 7


unless extremely tight and sophisticated (Garber, 1998). As regulations containingthe onshore use of derivatives just makes them move offshore, international co-operation between OECD and developing countries has to be fostered to bringoffshore markets under the umbrella of supervision.

The recent emerging market crises have brought the moral hazard ofcommercial and investment bank lending to developing countries to the forefront.The evidence from the Mexican and the Asian crises indicates that equity, long-termbond and local-currency investors have suffered heavy losses. By contrast, bankshave been the main beneficiaries of the Mexican bailout and have so far rarelysuffered large losses on their exposure to Asia (IMF, 1998a). This may also havedistorted the structure of capital flows from equity to debt, from long- to short-term and from local to foreign currency. While precise evidence on suchinteractions has not and cannot be established, recent events have shown thatmoral hazard of foreign investors cannot be eliminated cheaply.

The Russian unilateral moratorium, partly triggered by the refusal of theinternational community to provide further external assistance, was a clear firstwarning that “governments limit the scope and clarify the design of guarantees thatthey offer” (G22, 1998, p. 3). In the presence of modern risk-control methods(discussed above), the failure to bail Russia out once again has clearly spreadcontagion to other emerging markets (and beyond). Regardless of fundamentals,risk premia rose for all emerging markets. The rise may also have reflectedinvestors’ perception that the suggestions floated in the policy community on howto bail in creditors may actually raise default risk.

Some policy options to reduce the trade-off between exorcising moral hazardfrom foreign lending and intensifying crisis contagion exist, however. The G22suggests a standing, privately funded mechanism to provide new credits in the eventof a crisis or payments suspension. A prototype has been provided by Argentina andMexico, both of which have set up standby credit lines with banks which will lendin times of crisis. How these agreements work out in practice remains to be (stress)tested.

Other suggestions are all based on extending insurance or contingent-loanfacilities to countries which prequalify for such support by pursuing “sound”policies or by fulfilling defined banking standards. The attractiveness of such planslies in the restoration of a genuine lender of last resort for international assistance,since the club of potential beneficiaries and likelihood that they will succumb to

2 8


financial crises is reduced. Such pre-defined liquidity support might rapidly establishmarket confidence in times of panic. The important drawback of such proposals,however, is that they would cement the world into a class society where some cancount on support and others cannot. The experience of strings attached to liquiditysupport , designed to reduce moral hazard in official lending, has amply shown thatthe definition of eligibility for support becomes easily politicised.

A lot of attention has been given to the problem of how to correct theunderpricing of risk and hence excessive risk taking by banks. Relatively littlethought has been devoted to how to prevent crises by inducing a rise in stable flowsto developing countries. Long-term contractual savings institutions —pensionfunds and insurance companies — command a reliable long-term liability structureon their balance sheets and they pool large assets. These institutions have a greatcapacity to absorb risk, but regulation has often prevented them from investing inhigher-risk assets, such as those in below investment-grade emerging markets. Thereactive downgrading of Asian sovereign ratings from investment-grade to “junk”status reinforced the region’s crisis in many ways, e.g. by forcing institutionalinvestors to offload Asian assets as they were required to maintain portfolios onlyin investment-grade securities. Rating agencies have also reinforced global crisiscontagion in a self-fulfilling way by justifying the downgrading of Latin Americanassets with risks of contagion rather than fundamentals. The reactive (rather thanpreventive) approach of rating agencies can be explained by the fact that theinformation content of sovereign risk ratings and the nature of sovereign riskprovide only little room for rating agencies to acquire advance knowledge orsuperior information on emerging market economies (Reisen and von Maltzan,1998). Regulators should therefore reconsider the role of sovereign ratings thatthey stipulate when institutional investors hold emerging market assets. Theremoval of investment rating requirements might attenuate the boom-bust cyclein emerging-market lending by forcing banks and investors to rely on their ownjudgement rather than moving like herds on rating signals.

2 9


Crisis Prevention: What Host Countries Can Do

Private spending booms, fuelled by overborrowing, have increasingly led totwin banking and currency crises in developing countries, even in countries with areputation for sound macroeconomic fundamentals (IMF, 1998b). Private capitalinflows, attracted by exchange rate pegs and (often “disorderly”) financial opening,have repeatedly reinforced such pre-crisis booms. Domestic financial systems havetended to prove too weak as a conduit for heavy capital inflows, resulting indeclining credit quality and financial fragility (Reisen, 1998b)3.

The speculative currency attacks of the 1990s have challenged traditionalcrisis models that view them as a result of the government’s inability to achieve fiscaland monetary discipline. The vulnerability to attacks was driven by private bank andnon-bank borrowing, resulting in rising stock imbalances between real cashbalances, short-term debt and official reserves (Calvo and Mendoza, 1996), as wellas in currency and maturity mismatches. Once short-term foreign debt exceedsofficial reserves, a run on a country’s liquid assets is intensified by the investorknowledge that there are not enough liquid reserves to restore confidence(Radelet and Sachs, 1998). Short-term debt poses special problems for themaintenance of financial stability, as its rapid withdrawal can trigger sovereigndefault, a systemic banking and payments crisis and large-scale corporate defaults(Eichengreen, Mussa, et al., 1998).

In order to reap the benefits of global capital flows without falling victim totheir inherent risks, host countries have two broad policy sets at their disposal,macroeconomic discipline assumed. One set of policies serves to strengthen bankand non-bank balance sheets, so that capital inflows are intermediated and allocatedefficiently. The other set aims to raise the quality of these inflows. Neither of thesetwo policy sets, however, can be a substitute for the country’s strong determinationto maintain stability and to address instability should it arise.

Some of the avenues to strengthen a country’s balance sheets areuncontroversial; but they are deceptively hard to implement, in particular in thehistorical context of developing countries’ political, legal and institutionalbackgrounds. The recommendations of the G22 Working Groups on “Transparencyand Accountability” and “Strengthening Financial Systems” point to a long list ofingredients deemed critical for lessening the probability of financial imbalances:

3 0


— Good accounting standards and complete, accurate and timely informationdisclosure are a necessary precondition for prudential regulation andsupervision. Moreover, they can stabilise market expectations by improvingrisk assessment during the boom and by cushioning panic during a downturn,including crisis contagion. Transparency helps to promote accountability — by creating pressure on private and public decisionmakers to explain theiracts and to assume responsibility. Public policy can help create the appropriateenvironment by mandating the proper use of accounting, auditing andreporting rules. Moral hazard in the allocation of capital within countries,particularly relevant in the presence of connected and directed lending, hasto be exorcised by abolishing guarantees and through forcing banks and otherlenders to assume capital loss for ill-assessed credit risk.

— Only with reliable accounting systems and disclosure requirements to ensuretransparency will it be possible to strengthen bank and non-bank balance-sheets and to enforce prudential regulation through serious, independentsupervisory arrangements. It is safe to assume that basic ingredients foreffective enforcement of prudential regulation will meet resistance fromaffected interest groups. Nonetheless, the basic requirements are: independentinternal oversight of lending decisions by a credit review committee; vestingthe supervisory agency with the authority to examine bank operations andbalance sheets, close banks and establish entry criteria, define capital adequacyand exposure limits, enforce asset classification, provisioning rules andprudent collateral valuation that fully reflect the volatility of developing-country asset markets.

— The Asian crises, particularly in Indonesia and Korea, have also shown theimportance of sound practices in the area of corporate governance (Millstein,1998). More concerned with raising market share than with maximisingprofits, and reluctant to issue equity as this would dilute their managementcontrol, non-bank firms greatly contributed to overborrowing and currencymismatches by raising offshore short-term debt. This has created systemicrisk to entire economies, as large-scale default resulting from currencydevaluation threaten the stability of the banking system. Corporate riskmanagement and risk control, notably the management of liquidity and foreignexchange risk, are central to avoiding financial instability arising from accessto global markets.

3 1


These prescriptions, while largely uncontroversial and emphasised by officialthinking in both industrial and emerging countries, will take time to implement andwill be hard to maintain. In fact, the Asian crisis countries had tried to strengthenthe supervisory and regulatory infrastructure during the 1980s and 1990s, partly inresponse to costly banking crises (such as in Indonesia and Malaysia) a decade ago.What matters is the enforcement of prudential regulation.

Because the institutional capacity for durably strengthening balance sheets isnot built overnight, appropriate policies have to be formulated. Where therequired manpower is in short supply, allowing free entry by foreign banks andfinancial service providers can speed up capacity building. Importing rather thanbuilding expertise strengthens accounting practices, disclosure standards and riskmanagement practices as they are shaped by more demanding requirements in theforeign banks’ home countries. This, however, requires an orderly exit policy forailing domestic banks as foreign bank entry squeezes profit margins and encouragesailing domestic banks into high-risk bets for survival. It also requires free entry offinancial-sector experts and supervisors, as otherwise scarce supervisory personnelmay be drawn to the entrant banks, as happened in Thailand (Montes, 1998).

A second set of policies to dampen the risk from excessive short-term debt,less emphasised by official thinking, aims at raising the quality of capital inflows. Suchpolicies can be grouped under the headings of a) appropriate exchange ratemanagement, b) prudential and control measures to contain short-term flows intothe country, and c) promotion of long-term inflows, not least through orderlysequenced capital account liberalisation.

The excessive reliance on short-term borrowing can be discouraged byflexible exchange rates. By contrast, exchange rate pegs, in combination with highinterest rates, typical in developing countries for structural reasons, tend toreinforce bank lending and spending booms (Reisen, 1998a). They constitute anincentive for leveraged investors to exploit interest differentials as well as foroffshore borrowing by creditworthy banks and non-banks to tap seemingly cheapsources of finance. Central bank intervention on the foreign exchange market topeg the currency in the face of net inflows, unless sterilised fully, is intermediatedinto the domestic banking system. The exchange rate peg provides the incentive toallocate those funds disregarding currency and maturity risks, as these are beingimplicitly transferred to the central bank (Calvo and Mendoza, 1996). Keepingnominal exchange rates flexible, even introducing “noise” through central bankintervention when it is seen to be on a too-stable, appreciating trend during inflowperiods, improves the mix of inflows towards longer maturities and encouragesbanks and firms to hedge their foreign-currency exposures.

3 2


Box 1. Banco Central de Chile Studies on the Effectiveness of theCountry’s Measures to Regulate Capital Flows

The Chilean prudential framework, including capital inflow restrictions, has featuredprominently in policy discussion on how best to deal with volatile capital inflows. Chile’sauthorities followed two main policy targets in view of a surge in capital inflows during the1990s: First, to maintain a tight monetary policy without hindering export competitivenessresulting from unwarranted exchange rate appreciation. Second, to control the compositionof inflows by discouraging short-term capital so as to limit the short-term foreign debt andforeign currency exposure of both bank and non-bank entities. In 1991, the central bankimposed a one-year unremunerated reserve requirement (encaje) on foreign loans.Subsequently, the rate of the encaje was increased to 30 per cent and its coverage extendedto cover virtually all foreign inflows except foreign direct investment. The one-year minimumholding period effectively implied a tax on inflows, inversely tied to their maturity. Prudentialregulation complemented these curbs on inflows. Except for trade credits, banks cannot lenddomestically in foreign currency. And maximum open foreign exchange positions are set at20 per cent of banks’ capital and reserves. As short-term flows dried up in 1998, the encajewas reduced to 0 per cent, but not abolished.As it is difficult to quantify the intensity of inflow restrictions and to control for prudentialregulations, macroeconomic policies and other conditions that impact on capital inflows, theeffectiveness of Chile’s measures is being hotly debated. The Banco Central de Chile hasprovided the OECD Development Centre with a set of unpublished internal studies on theimpact of the encaje1. These authoritative studies do show that the encaje has been effectivein providing some monetary autonomy and in influencing the mix of capital inflows. Eyzaguirreand Schmidt-Hebbel (1997) set up a model for analysing and estimating the dynamic effectsof the encaje on capital inflows. The model predicts an intensification of the encaje to resultin higher domestic interest rates, diminished net foreign debt and a depreciation of theexchange rate. These predictions are borne out by calibrating the model with monthly datafor the period January 1991 to June 1996. It is also shown that the encaje, with a lag of oneyear, modifies the composition of inflows by reducing the share of short-term flows in favourof longer maturities. The paper does not explore, however, to what extent the improved mixof inflows represents relabeling in order to escape the implicit tax on short-term flows. LeFort and Sanhueza (1997) also provide evidence for 1990-96 that Chilean capital controlshave been effective in keeping domestic interest rates above international rates, and that theeffectiveness has not been eroded over time. Moreover, each time the coverage of the encajehas been extended, the newly taxed inflows have been reduced without a full substitutiontowards tax-free flow items. However, while the study is empirical, it does not provide aneconometric analysis of the degree of effectiveness. Soto (1997) runs a vector autoregressionanalysis on capital flows, interest rates and the real exchange rate for June 1991 to June 1996.He finds that capital controls have the desired effect of reducing capital inflows, maintaininghigher interest rates and a lower real exchange rate, and reducing the share of short-termcapital inflows. However, the magnitude of these effects is fairly small. One explanation forthe small impact of the encaje on the dependent variable may be that the implicit tax of theinflow controls are smaller than assumed in most studies. Considering the positive optionvalue of closing the investment position or staying invested in Chile once the investment hasbeen done, reduces the implicit tax on a one-year investment from 2.50 to 1.25 per cent(Herrera and Valdés, 1997).

1. The author gratefully acknowledges the co-operation of Klaus Schmidt-Hebbel, Headof Research at the Banco Central de Chile.

3 3


Most mutual funds, pension funds and life insurers impose penalties for earlywithdrawal by investors. Chile, with its one-year unremunerated reserve requirementgradually extended to all inflows except foreign direct investment, has donelikewise, as has Colombia (Ffrench-Davis and Reisen, 1998). Such capital inflowrestrictions provide policymakers with a policy instrument in the policy trilemmathat free capital flows, an independent monetary policy and pegged exchange ratesare mutually inconsistent. They also extend the range of prudential regulationmeasures to limit the upwinding and unwinding of foreign-currency short-termpositions, which have devastated emerging-market banking systems and economiesin the past, including Chile’s in 1982.

It has been argued that Chile’s reserve requirements amount to a distortionthat does not allow capital to flow to uses that offer the highest rate of return. Tothe extent, however, that these flows pose an exogenous distortion to returns(e.g., when high flows with multiple leverage drive asset prices up and down) or thatthe structure of foreign capital supply is distorted (e.g., by the Basle adequacyregimes, discussed above) capital restrictions on short-term inflows can be seen ascorrecting rather than creating a distortion. Finally, when the short-term debt/reserves ratio dangerously approaches unity, avoiding a speculative attack impliesthe need for the central bank to put every dollar of increased debt into officialreserves to prevent the vulnerability ratio from growing. For developing countries,this means borrowing at a higher rate of interest than the rate at which funds arereinvested, say, into US Treasury Bills. Such a swap clearly constitutes a negativeexternality to the country, hence the rational to restrict short-term borrowing atthe source.

Are caps on short-term inflows effective in improving the structure ofinflows? After all, the high degree of integration in trade, production and financialservices opens up many ways of circumventing controls. One reason that capitalcontrols in most OECD countries were abolished in the 1980s was the perceptionthat they were increasingly ineffective. Restrictions on inflows, but also prudentialregulations on open foreign currency positions, are difficult to enforce as banks canuse offshore subsidiaries or derivatives to evade them (Garber, 1998)4. Authoritativestudies from Chile’s Central Bank do show that Chile’s measures to regulate capitalflows have been effective in providing a degree of monetary autonomy and ininfluencing the size and mix of capital inflows (Box 1). But the impact has been weak,and regulatory measures in Chile have been supported by a culture of transparencyand enforcement as well as by a set of macroeconomic policies (balanced budgets,wide target zones for exchange rates) consistent with raising the share of long-terminflows (Ffrench-Davis and Reisen, 1998). These conditions have not always beenpresent in other emerging economies.

3 4


While Chile’s controls were designed to tax short-term inflows withouthindering long-term portfolio and direct investment, financial opening in the worst-hit Asian economies was “disorderly” (Poret, 1998; Reisen, 1998b). Often theresult of discretionary authorisation granted to selected sectors, the openingprocess implicitly encouraged short-term inflows in the pre-crisis years. While tightquantitative ceilings were maintained on non-resident purchases on the stockmarket and on foreign direct investment (notably in Korea), financial opening easedaccess to short-term foreign borrowing. Table 11 may be indicative of the lessonsto be drawn: the rise in short-term debt can be contained through “orderly”liberalisation (as in Colombia and Chile); but disorderly liberalisation (as in theAsian crisis countries) encourages financial vulnerability, in particular if a stronglyenforced supervisory framework is not yet in place.

Table 11. Maturity Structure of Foreign Debt in Selected Countriesper cent of short-term in foreign debt, end-June 1997

Country Short-term Debt

Colombia 39.4Chile 43.3

Malaysia 56.4Philippines 58.8Indonesia 59.0Thailand 65.7Korea 67.9

Source: Bank for International Settlements, The Maturity, Sectoral and Nationality Distribution of International Bank Lending,Basle, May 1988.

These observations reinforce the need for a capacity-building sequence ofliberalising capital inflows, as has indeed been advocated since the early 1990s(Fischer and Reisen, 1992). Foreign direct investment and trade-related finance,while a necessary ingredient for development even at the earliest stage, are unlikelyto cause trouble for macroeconomic management and financial sector stability.They are early candidates for liberalisation, while other capital flows confront theauthorities with more complicated issues. In view of the considerable time neededto establish a sound domestic financial system — accounting, auditing, disclosure,regulation and supervision — the required infrastructure should be built withoutdelay and be enhanced by liberalising the entry of financial-market expertise. Thisrequires a clear and durable solution of prior bad-loan problems in the bankingsector. The next candidates for liberalisation are portfolio equity and long-termbond investments, which should be fostered in parallel with building the infrastructurefor domestic stock, corporate debt and mortgage instruments. This will deepen

3 5


domestic money markets, which allow authorities to smooth shocks to domesticliquidity. Deepened domestic financial markets pave the way for dismantlingcontrols on short-term borrowing by banks and non-banks, assuming a toughsupervisory regime is in place. Financial opening has the best chance to achieve itsultimate objective, to raise efficiency and growth without compromising stability,when combining a sequential opening process with building the prerequisiteinstitutions.

Notes

1. Capital inflows can be shown to be “immiserising” (Brecher and Diaz-Alejandro, 1997), if theymagnify welfare losses due to distorted consumption and production patterns by stimulatingcapital accumulation in protected sectors and by attracting foreign capital into these sectors.

2. The impact of hedge funds on the Asian crisis is difficult to determine with any precision.Eichengreen and Mathieson (1998) find little evidence that hedge funds led the crisis. The BIS(1998), by contrast, reports strong credit demand from offshore financial centres (wherethese funds are mostly incorporated). Market participants (Garber, 1998; Howell, 1998)maintain that the absence of data on hedge fund transactions, such as Over-The-Counter(OTC) contracts written by investment banks, clearly leads to underestimation of theirimpact.

3. To add another narrative to the large literature on the domestic causes of the Asian crisis,is beyond the scope of this Brief. See in particular IMF (1998a), Davies (1998), Corsetti et al.(1998), Radelet and Sachs (1998) and Reisen (1998b).

4. The BIS (1998) suggests in its 1998 annual report that the withdrawal of rights of establishmentfor banks from jurisdictions with inadequate prudential standards might have to be contemplated.

3 6


Bibliography

ARMSTRONG, A. AND M. SPENCER (1998), “Will the Asian Phoenix Rise Again?”, Global EmergingMarkets, Vol. 1.3, Deutsche Bank Research, London.

BANK FOR INTERNATIONAL SETTLEMENTS (BIS)(1998), 68th Annual Report 1997/98.

BHAGWATI, J. (1998), “The Capital Myth”, Foreign Affairs, Vol. 77.3.

BORENSZTEIN, E., J. DE GREGORIO AND J.-W. LEE (1995), “How Does Foreign Direct InvestmentAffect Economic Growth?”, NBER Working Paper No. 5057.

BRECHER, R.A. AND C.F. DIAZ-ALEJANDRO (1997), “Tariffs, Foreign Capital and ImmiserizingGrowth”, Journal of International Economics, Vol. 7.4.

CALVO, G. AND E. MENDOZA (1996), “Petty Crime and Cruel Punishment: Lessons from theMexican Debacle”, American Economic Review, Papers and Proceedings.

CLINE, W. (1995), International Debt Reexamined, Institute for International Economics,Washington, D.C.

CORSETTI, G., P. PESENTI AND N. ROUBINI (1998), “What Caused the Asian Currency andFinancial Crisis?”, mimeo (http://www.stern.nyu.edu/~nroubini/asia/AsiaHomepage.html).

D’ANDREA TYSON, L. (1998), “The Global Meltdown: What To Do Next”, Business Week,October 26.

DAVIES, G. (1998), Causes, Cures and Consequences of the Asian Economic Crisis , GoldmanSachs, London.

DOOLEY, M. (1994), “A Retrospective on the Debt Crisis, NBER Working Paper No. 4963.

DOOLEY, M. (1995), “A Survey of Academic Literature on Controls over InternationalCapital Transactions”, NBER Working Paper No. 5352.

EICHENGREEN, B. AND D. MATHIESON (1998), “Hedge Funds and Financial Market Dynamics”,IMF Occasional Paper No. 166.

EICHENGREEN, B., M. MUSSA et al. (1998), “Capital Account Liberalization: Theoretical andPractical Aspects”, IMF Occasional Paper No. 172.

3 7


EYZAGUIRRE, N. AND K. SCHMIDT-HEBBEL (1997), “Encaje a la entrada de capitales y ajustemacroeconomico”, Banco Central de Chile, mimeo.

FFRENCH-DAVIS R. AND H. REISEN (1998), Capital Flows and Investment Performance: Lessons fromLatin America, ECLAC, Santiago and OECD Development Centre, Paris.

FISCHER, B. AND H. REISEN (1992), “Towards Capital Account Convertibility”, Policy BriefNo. 4, OECD Development Centre.

FOLKERTS-LANDAU, D. AND P. GARBER (1998), “Capital Flows From Emerging Markets in aClosing Environment”, Global Emerging Markets, Vol. 1.3, Deutsche Bank Research,London.

FRANKEL, J. (1994), “Sterilization of Money Inflows: Difficult (Calvo) or Easy (Reisen)?”, IMFWorking Paper WP/94/159.

GARBER, P. (1998), “Derivatives in International Capital Flows”, NBER Working PaperNo. 6623.

GREENSPAN, A. (1998), Remarks by Alan Greenspan, Chairman, Board of Governors of the FederalReserve System, at the 34th Annual Conference on Bank Structure and Competitionof the Federal Reserve Bank of Chicago, May 7.

GRILLI, V. AND G. MILESI-FERRETTI (1995), “Economic Effects and Structural Determinants ofCapital Controls”, IMF Staff Papers, Vol. 42.3.

GROUP OF TEN (1996), “The Resolution of Sovereign Liquidity Crises: A Report to theMinisters and Governors”, Basle.

GROUP OF TWENTY-TWO (1998), “Summary of Reports on the International FinancialArchitecture”, Washington, D.C.

HERRERA, L. AND R. VALDÉS (1997), “Encaje y autonomia monetaria en Chile”, Banco Centralde Chile, mimeo.

HOWELL, M. (1998), Asia’s ‘Victorian’ Financial Crisis, Cross Border Capital, London.

ILO (1998), The Social Impact of the Asian Financial Crisis, ILO, Geneva.

IMF (1998a), International Capital Markets: Development, Prospects and Key Policy Issues,Washington, D.C.

IMF (1998b), World Economic Outlook, Washington, D.C.

IZE, A. AND G. ORTIZ (1987), “Fiscal Rigidities, Public Debt and Capital Flight”, IMF StaffPapers, Vol. 34.2.

KENEN, P. (1993), “Financial Opening and the Exchange Rate Regime”, in H. REISEN AND

B. FISCHER (eds.), Financial Opening: Policy Issues and Experiences in Developing Countries,OECD, Paris.

3 8


KINDLEBERGER, C. (1978), Manias, Panics and Crashes, Basic Books, New York.

KRUGMAN, P. (1998), “Saving Asia: It’s Time to Get Radical”, Fortune, September 7.

LE FORT, G. AND G. SANHUEZA (1997), “Flujo de capitales y encaja en la experienza chilenade los 90”, Banco Central de Chile, mimeo.

MACKELLAR, L. AND H. REISEN (1998), “A Simulation Model of Global Pension FundInvestment”,Technical Paper No. 137, OECD Development Centre; forthcoming,Journal of Economic Policy Modelling.

MILLSTEIN, I. (1998), “Corporate Governance: Improving Competitiveness and Access toCapital in Global Markets — A Report to the OECD”, OECD, Business SectorAdvisory Group on Corporate Governance, Paris.

MONTES, M. (forthcoming), “The Asian Economic Crisis: What is to Be Done?”, OECDDevelopment Centre Technical Paper.

OECD (1997), The World in 2020: Towards a New Global Age, Paris.

OECD (1998), OECD Economic Outlook, No. 63, Paris.

PORET, P. (1998), “Liberalising Capital Flows: Lessons from Asia”, OECD Observer, No. 214.

POSEN, A. (1998), Restoring Japan’s Economic Growth, Institute for International Economics,Washington, D.C.

RADELET, S. AND J. SACHS (1998), “The Onset of the East Asian Financial Crisis”, mimeo (http://www.hiid.harvard.edu/).

REGLING, K. (1999), “Financial Liberalisation in Asia: Analysis and Prospects”, in M. QUEISSER

(ed.), Financial Liberalisation in Asia: Analysis and Prospects , ADB/OECD.

REISEN, H. (1989), “Public Debt, External Competitiveness, and Fiscal Discipline in DevelopingCountries”, Princeton Studies in International Finance No. 66.

REISEN, H. (1994), “The Brady Plan and Adjustment Incentives”, in H. REISEN, Debt, Deficitsand Exchange Rates: Essays on Financial Interdependence and Development , Edward ElgarPublishing Ltd., Aldershot, (UK).

REISEN, H. (1996), “Managing Volatile Capital Inflows: The Experience of the 1990s”, AsianDevelopment Review, Vol. 14.1.

REISEN, H. (1998a), “Excessive and Sustainable Current Account Deficits”, Technical PaperNo. 132, OECD Development Centre, published in Empirica, Vol. 25.2.

REISEN, H. (1998b), “Domestic Causes of Currency Crises: Policy Lessons for CrisisAvoidance”, Technical Paper No. 136, OECD Development Centre.

REISEN, H. AND A. VAN TROTSENBURG (1988), Developing Country Debt: The Budgetary andTransfer Problem, OECD Development Centre Studies, Paris.

3 9


REISEN, H. AND J. VON MALTZAN (1999), “Boom and Bust and Sovereign Ratings”, InternationalFinance, Vol. 1.3.

RIVLIN, A. (1998), “Lessons Drawn from the Asian Financial Crisis”, reprinted in BIS ReviewNo. 41, 13 May.

RODRIK, D. (1998), “Who Needs Capital Account Convertibility?”, in P. KENEN (ed.), Shouldthe IMF Pursue Capital Account Convertibility?, Princeton Essays in International Finance,No. 207.

SOTO, C. (1997), “Controles a los movimentos de capital: evaluación empirica del casochileno”, Banco Central de Chile, mimeo.

SUMMERS, L. (1998), “Building an International Financial Architecture for the 21st Century”,Remarks to the Cato Institute, 22 October.

WORLD BANK (1998), East Asia’s Road to Recovery, Washington, D.C.

WYPLOSZ, C. (1998), “International Financial Instability”, Graduate Institute for InternationalStudies, Geneva, mimeo.

4 0


Other Titles in the Series

Adjustment and Equity (No. 1)by Christian Morrisson (January 1992)

Managing the Environment in Developing Countries (No. 2)by David O'Connor and David Turnham (January 1992)

Privatisation in Developing Countries: Reflections on a Panacea (No. 3)by Olivier Bouin (April 1992)

Towards Capital Account Convertibility (No. 4)by Bernhard Fischer and Helmut Reisen (April 1992)

Trade Liberalisation: What's at Stake? (No. 5)by Ian Goldin and Dominique van der Mensbrugghe (May 1992)

Towards Sustainable Development in Rural Africa (No. 6)by David Turnham, with Leif E. Christoffersen and J. Thomas Hexner (January 1993)

Employment Creation and Development Strategy (No. 7)by David Turnham (July 1993)

The Disarmament Dividend: Challenges for Development Policy (No. 8)by Jean-Claude Berthélemy, Robert S. McNamara and Somnath Sen (April 1994)

Pension Fund Investment from Ageing to Emerging Markets (No. 9)by Bernhard Fischer and Helmut Reisen (January 1995)

What Institutional Framework for the Informal Sector? (No. 10)by Christian Morrisson (October 1995)

The Policy Challenges of Globalisation and Regionalisation (No. 11)by Charles Oman (June 1996)

Policies for Economic Take-off (No. 12)by Jean-Claude Berthélemy and Aristomène Varoudakis (September 1996)

The Political Feasibility of Adjustment (No. 13)by Christian Morrisson (October 1996)

Biotechnology Policy for Developing Country Agriculture (No. 14)by Carliene Brenner (April 1997)

Pension Reform: Lessons from Latin America (No. 15)by Monika Queisser (January 1999)

Documents

OECD DEVELOPMENT CENTRE · 2016. 3. 29. · OECD DEVELOPMENT CENTRE 94 rue Chardon-Lagache 75016 Paris, France Tel.: (33-01) 45.24.82.00 Fax: (33-01) 45.24.79.43 DEVELOPMENT CENTRE